| 摘要 |
<p>The identification of hidden data, such as feature-based control points in an image, from a set of observable data, such as the image, is achieved through a two-stage approach. The first stage involves a learning process, in which a number of sample data sets, e.g. images, are analyzed to identify the correspondence between observable data, such as visual aspects of the image, and the desired hidden data, such as the control points. Two models are created. A feature appearance-only model is created from aligned examples of the feature in the observed data. In addition, each labeled data set is processed to generate a coupled model of the aligned observed data and the associated hidden data. In the image processing embodiment, these two models might be affine manifold models of an object's appearance and of the coupling between that appearance and a set of locations of the object's surface. In the second stage of the process, the modeled feature is located in an unmarked, unaligned data set, using the feature appearance-only model. This location is used as an alignment point and the coupled model is then applied to the aligned data, giving an estimate of the hidden data values for that data set. In the image processing example, the object's appearance model is compared to different image locations. The matching locations are then used as alignment points for estimating the locations on the object's surface from the appearance in that aligned image and from the coupled model.</p> |
